This year alone, America has been devastated by 47 different mass shootings. In response to the recent Florida tragedy, our team discussed ways to utilize technology in order to help the people involved in these situations. Our thoughts immediately went to first responders and those who bravely help others during these events. We thought that in the crucial 5 to 10 minutes after a victim is shot, even someone with little or no CPR experience should be armed with the resources to save them. And with that realization, we set out to develop our app.

As we thought through different uses for this technology, I recalled my own CPR certification from my summers as a camp counselor. I have a card that says I am certified, but if a child needed CPR, I don't know that I would be capable of administering it. With this technology, we wanted also to give people who may not remember CPR protocols in the heat of the moment a way to easily deliver CPR as well.

What it does

Our app is intended to guide users through CPR on the person they are trying to save. As it is opened, the users take a picture of a body, then the app uses image processing in order to recognize where the middle of the person's chest and mouth are. With this knowledge, the app goes through step-by-step narrated instructions complete with animations and necessary sound effects (the beat at the rate which hands should be compressing, rescue breaths etc) on the body itself. This is essential for people who may not remember exactly how or where to do necessary CPR motions. Since the animations point out precisely where to perform the CPR steps on the victims themselves, the possibility of forgetting or being too wrapped up in the moment to deliver CPR correctly is greatly diminished. We also wanted to include a place emergency phone call function directly in the app so that people using it can easily call emergency personnel.

How we built it

We utilized Android Studio to create the app portions and take the pictures. For 3D animations (which we could not include in our demo), we used Blender. Since these were not ready for use just yet, we instead placed 2D animations in their spots using Photoshop and Premiere. To do the face and chest tracking we used a javascript library called clmtrackr. We used Audacity to process the narration and other sound effects.

Challenges we ran into

The first thing we had trouble with was our implementing our idea. We were all in love with the concept of aiding people with CPR, but we didn't know the best way to do that. Logistically, trying to find the best way to tackle this problem in a way that would be useful to users (and help the greatest number of people in the greatest number of situations) was difficult for us and ended up taking up most of our time on Saturday. We also couldn't commit to one type of technology, imagining our idea in AR, VR, as a website and eventually settling on an app. This was not only time consuming, but also physically and emotionally draining (if you need proof, you can look at the 4 whiteboards we took up). One of our team members put it perfectly, we just had analysis paralysis. Once we managed to break free and actually start building our concept, we were challenged by all sorts of new softwares. This was definitely the most frustrating and enjoyable part of our process. Everyone on our team had the chance (and challenge) of learning a software that was entirely new to them. The last challenge we had (which is part of what makes hacakthons so awesome) was time. We all attended workshops (which we don't regret in the least!), but the time we spent at workshops was time we could not use to develop our app. In the end, we had to scale back on our original plan (at least to submit) because we didn't have enough time to finish all that we wanted to accomplish.

Accomplishments that we're proud of

The biggest thing that we are proud of is that we all still like each other. That sounds sort of odd, but we think it is a huge accomplishment. We all had vastly different plans but managed to disagree constructively and work together well as a team. We spent many hours revising our plan (as you can see above), each person bringing their own ideas, creativity, and brilliance to the table to make our hack a success. The fact that we had never known each other before this hackathon then managed to spend 24 hours together without killing one another (except for the purposes of testing face recognition for CPR) and actually come out close friends is awesome. Another thing we are proud of is figuring out how to creatively find the middle of a user's chest. We discovered (then tested) a theory which said that the height of one's head is equal to the distance between the chin and middle of their chest. Again, not a humongous achievement technically, but it was an issue which stumped us for a good amount of time and we were extremely proud when we worked together to figure out a solution. Technically, we are proud that we have something to showcase. Since the software was all new to each of us, the fact that we were able to create anything at all with it is an accomplishment.

What we learned

Oh my gosh...where to even start? We definitely learned about each other and even more about how to work together. Before 11 on Saturday, none of us had ever even met, let alone built a project together into the middle of the night. That is one thing I personally love about hackathons! Anyway, we had to figure out how to harness our own strengths and weaknesses and unite to build an app we all liked. Fortunately, this was not difficult and we definitely shared a lot of laughs. We also all learned TONS of things technically as well. We learned how to make animations on blender, how to use android studio, how face detecting algorithms work, how much we hate how git hub works(just kidding, but not really!) Most of all we learned how much we have yet to learn.

What's next for How to Save a Life

First, we need to connect all of the components of our project together. We all worked on different aspects of this idea and need to flesh them out when we are running on more sleep. We also want to expand our app to include options for children or infants. After that, there are infinite opportunities. We could bring this app to camps for counselors or provide it to teachers or others who work with those who may need CPR. In addition, it could be used for training first responders or those who want to learn CPR quickly and easily. This could be marketed to people who already know first aid, but may not be confident in their abilities.

Built With

Share this project: